Although cancer-promoting activities of extracellular proteases and peptidases in the tumor microenvironment are well known, the specific disruption of a single enzyme in basic and translational settings remains a fundamental challenge. For example, the use of pan-specific chemical inhibitors of matrix metalloproteinases (MMPs) in animal models of cancer has demonstrated that interfering with MMP activity can slow cancer progression, but clinical translation of these inhibitors has failed, largely due to lack of inhibitor specificity. Efforts to discover specific inhibitors or rapidly implement genetic approaches to interfere with MMPs and other enzymes active in the tumor microenvironment remain challenging. Emerging findings indicate that MMPs modulate signaling and immune function in the tumor microenvironment, making highly specific reagents critical to elucidate the wide range of roles MMPs play within the tumor microenvironment and to establish a new generation of inhibitors for potential clinical applications. In this application, we propose to establish a platform for the discovery of highly specific enzyme inhibitors in high throughput. Our underlying technology combines the use of yeast display and noncanonical amino acids to construct and evaluate inhibitors with structures that are not accessible using conventional approaches. We hypothesize that precisely positioning small molecules within an antibody framework will yield bivalent ?hybrid inhibitors? that retain antigen specificity while gaining potent inhibitory capabilities. We will implement our yeast-based discovery platform by targeting MMP-2, -7, -9, and -14 because of the availability of numerous MMP-related tools that will allow us to explore the capabilities of our platform. We will establish our platform using the following two specific aims: 1) Identify effective small molecule positioning in antibodies to enable efficient hybrid discovery. In this Aim, we will quantitatively explore the effects of small molecule attachment sites, functional groups, and linkers within antibody variable regions to identify the most promising combinations of these factors for large-scale hybrid discovery. 2) Establish quantitative relationships between inhibitor properties and cell invasion. In this Aim, we will compare data we derive from yeast-based measurements with data from standard biochemical and cell-based assays in order to set numerical targets during hybrid discovery efforts and to examine the roles of MMPs in cell invasion. The platform proposed here will yield molecular reagents with structures and specificities that cannot be accessed using existing small molecule- or protein-based approaches. The resulting MMP inhibitors may serve as therapeutic leads, as their high specificities will overcome key shortcomings of previous therapeutic candidates. Precise enzyme disruption in the tumor microenvironment will also lead to a better understanding of tumor biology at both molecular and systems levels. Finally, we anticipate that the hybrid approach to enzyme inhibition will be applicable to member-specific targeting of any family of enzymes.